Epistemological Considerations

administered survey instrument is part-strategy to eliminate the researcher’s bias and to enable survey participants’ to respond to the questions put to them without the intervention of the researcher or his assistants during the survey administration phase. This approach allows to a degree, time and context free generalizations to be made about responses (Nagel, 1986).

“the external world consists of pre-existing hard tangible structures which exist independently of an individual’s cognition” (Fitzgerald & Howcraft, 1998, p.323). Positivism is a product of the natural sciences, portrayed by the testing of hypotheses formulated from extant theories to measure social reality. The positivist believes that a scientific approach towards the advancement of knowledge ensures validity, confidence, and precision (Crotty, 1998 p.29). Said differently, the positivist paradigm is premised on repeatability, reductionism and reliability (Pather & Remenyi, 2004). Further, the positivist reckons, consistent with the realist ontology, that the researcher and the researched are sovereign (Lincoln & Guba, 2000a). This links to the objectivist standpoint: the researcher and the phenomena under investigation must be autonomous whereby the researcher remains neutral (Fitzgerald & Howcroft, 1998). Any interference during investigations could threaten the scientific validity of the results. Additionally, propositions must be logically true or empirically testable (Landry & Banville, 1992). Therefore, the focus of the positivist is on explanation associated with prediction and then control of the researched phenomenon (Orlikowski & Baroudi, 1991), and the realist on value free and explicit interpretation of reality (Guba & Lincoln, 1994).

4.1.2.2 Interpretivist and Subjectivist

Likewise the positivist and objectivist view, the interpretivist and subjectivist views are intertwined. These views complement each other as they both hold that the researcher and the researched are not sovereign. The interpretivist researcher subscribes to the relativist ontology, professing that: “multiple realities exist as subjective constructions of the mind, where socially- transmitted terms direct how reality is perceived and this will vary across different languages and cultures” (Fitzgerald & Howcraft, 1998, p.325). As such, the interpretivist concept is informed by a concern to identify with the world as it is, comprehend the primary nature of the social world at the level of subjective experience, and seek explanation within the boundaries of individual perception and subjectivity (Burrell & Morgan, 1979, p.28). In essence, scholars whom identify with this paradigm argue that reality can only be understood by subjective analysis of a phenomenon and intervention in reality (Klein & Myers, 1999). The interpretivist paradigm therefore seeks enlightenment through the connotations that individuals enact to them (Boland, 1985; Orlikowski & Baroudi, 1991; Deetz, 1996). Last, the interpretivist paradigm supports subjective descriptions and understanding, above the explanation and prediction goals linked with positivist research (Nissen, 1985).

4.1.2.3 Critical Theory

Conceptually, critical theory was built on the foundation of Marxism, by scholars who believed that traditional Marxism would not be an applicable theory for modern society to address its complex social and economic structures (Willis, 2007). The critical theory paradigm holds an ontological view that emphasizes a pensive assessment and critique of society and culture through the lens of social sciences and humanities (Sim & Van Loon, 2001). In essence, this is a reality that is created and profiled by social, economic, political, and cultural factors that have been fashioned eventually (Guba & Lincoln, 1994). As such, a critical theorist believes in a single reality that is shaped in time by social drivers. Further, Guba and Lincoln (1994) opine that the critical theorists are a closer fit with the subjectivist epistemology, in that; the researcher and inquiry are not sovereign. Methodological protocols adopted by critical theorists are often engagement methods such as: conversations and reflections to contend assumptions. In précis, Giroux, (1988, p.213) writes that the goal of engagement between the researcher and research participants’ is to identify various forms of past and dominated knowledge that reference experiences of affliction, divergence, and collective struggle, and to relate the idea of past understanding to essentials of critique and optimism. Thus, the goal of the critical theorist is to evaluate, alter, and liberate the social reality under scrutiny (Orlikowski & Baroudi, 1991;

Hirschheim & Klein, 1994; Klein & Myers, 1999; Myers & Klein, 2011). For this study, critical theory is not a fit because the study aims to quantify user perception towards M-pesa, rather than alter and change the research subjects’ worldview or critique society in view of experiences and ideas of past understanding.

4.1.2.4 Post-Positivism

The disparity between positivism and interpretivism has been largely debated (Fitzgerald &

Howcroft, 1998). In attempt to reconcile differences between these paradigms, Hirschheim (1985) proposes a post-positivism paradigm. Post-positivism presents a slightly modified ontological view to the realist and terms it ‘critical realism.’ The critical realism ontology claims to ascend the raw realism held by the positivist scholars, and rather posits that reality can be accepted only defectively and prospectively (Lincoln & Guba, 2000b, p.168). In essence, discoveries are identified by the researcher’s sentiments and acuity which dispels the grasp of a pure reality (Teddlie & Tashakkori, 2009).

The post-positivist doctrine promotes methodological pluralism, implying that a single method of science is erroneous; rather multiple methods strengthen our findings (Wildemuth, 1993). Chen and Hirschheim (2004) concur with the view that pluralism fosters the body of knowledge;

engaging substitute approaches for research strengthens or refines extant knowledge, an essential for development within the IS discipline. Post-positivism, although a commendable paradigm for its claim to reduce errors in findings, the researcher grapples with positivistic instinct to criticize the vagueness of language (i.e., defectively and prospectively) that conjoins paradigms, beliefs, and disciplines (Biersteker. 1989). In essence, concurring with Biersteker (1989) and by no intent a whippersnapper, the researcher holds cynicism about scholarship that professes indistinct criteria for selection amongst the various and sometimes conflicting explanations it yields, and believes that post-positivism is an irresolute approach to answering the conclusive based RQ’s posed in this study.

4.1.2.5 The Researcher’s Epistemological Stance

“Understanding in terms of cause and effect was an a priori characteristic of the human mind underlying all human knowledge.” ~ (Gregor, 2006, p.617 citing Kant 1781)

4.1.2.5.1 Discharged Views

The interpretivist, subjectivist, critical theorist, and post-positivist views are discharged as perspectives from this study because they do not maintain a clear distinction between science, personal experience, and fact and value judgement.

Interpretivist and subjectivist approaches attempt to understand a phenomenon based on relativity;

research subjects assigned meaning to them (Orlikowski & Baroudi, 1991; Deetz, 1996), critical theorists seek to assess, modify, and liberate the social reality under scrutiny (Orlikowski &

Baroudi, 1991), and post-positivist make discoveries through the researcher’s view and perception (Teddlie & Tashakkori, 2009), and do not generalize causes and effect (Neuman, 2000; Hudson & Ozanne, 1988). Collectively, these views favour descriptions and understanding over the explanation and prediction goals associated with positivist research (Nissen, 1985).

Views of these natures cannot concisely answer the cause and effect/conclusive based questions posed in this study:

(i) What are the antecedents of user continuance intention towards M-pesa?

(ii) What are the determinants of user continuance intention towards M-pesa?

To answer the afore-noted questions, the positivist view seems appropriate because it holds that the world of phenomena has an objective reality that may be reported in causal relationships and measured in data, in a reportable and accurate manner (Kaplan & Duchon, 1988; Straub, Boudreau, & Gefen, 2004). Essentially, causality or conclusive-based questions like the abovementioned can be investigated within an objective reality to enable the researcher report underlying associations, determine the associations in data in an accountable and truthful manner.

Elaboration on this view is presented next.

4.1.2.5.2 Selected View

In light of earlier discussions of overlaps between positivist and objectivist views in section 4.1.2.1, the researcher submits to the concept of an objective reality, and specifies that the epistemological viewpoint is positivist. It is thus viewed as the researcher’s investigation of the underlying laws of causation of a user’s continuance intention towards M-pesa. Aforesaid, the positivist paradigm applies a systematic, scientific approach to research, with a worldview of established laws of causation; everything that occurs around us can be explained by knowledge of the underlying universal laws (Fitzgerald & Howcroft, 1998). The goal of this study is to answer the RQ’s: (i) what are the antecedents of user continuance intention towards M-pesa, (ii) what are the determinants of user continuance intention towards M-pesa. To answer these questions, positivism holds that there are universal laws governing continuance intention, and to understand these universal laws, the researcher needs to observe and document events of the phenomena around the research setting in a systematic way and then work out the underlying principle that causes users to continue use of M-pesa. This approach aligns with Kant’s (1781) view that the world of phenomena has an objective reality, which can be conveyed in causal relationships and measured in data in a representative and precise method (Kaplan & Duchon, 1988; Straub et al.

2004). The core of a positivist research is to uncover the objective physical and social reality by modelling specific measures that will identify the dimensions of reality sought by the researcher (Orlikowski & Baroudi, 1991). A priori fixed relationships within a given phenomenon are often examined with structured instruments to validate theory, and this helps to advance predictive knowledge of a phenomenon (Orlikowski & Baroudi, 1991). The positivist doctrine asserts that a given phenomenon is structured and exhibits constancy, and thus, the researcher’s goal is to identify the patterns and consistencies, and then report them in form of causation (Cecez-

Kecmanovic, 2005). As such, causal laws are instructive in the control and prediction of technology use, user behaviour and attitudes towards a technology (Cecez-Kecmanovic, 2005).

4.1.2.5.3 Application of Selected View

Consistent with the above discussions, to convey the phenomenon of user continuance intention towards M-pesa in causal relationships, the researcher has identified suggested antecedents and determinants within the extant literature on consumer behaviour, computing behaviour, and psychology. The suggested factors have been adopted and modelled for statistical tests. Specific measures (constructs) to identify the dimensions of reality have been adopted from the extant studies, and data will be collected from users of M-pesa with survey instruments containing items measured on a likert scale. Thereafter, to examine a priori fixed relationships, quantitative methods of multivariate data are employed to assess survey participants’ responses. The results will aid the researcher in revealing the patterns and consistencies of user continuance intention towards M-pesa in a reported form of causation.

In sum, given this study’s goal to identify the underlying causal relationships that can predict user continuance intention towards M-pesa, and that the study has constructed an empirically testable model to determine causal relationships amongst the study’s predictor variables (perceived TTF, post-usage usefulness, confirmation, system quality, information quality, and service quality, utilization, satisfaction, flow, and trust, utilization, satisfaction, and trust), and the criterion variable (continuance intention), the characteristics of the positivist approach is the most appropriate for the execution of the study.